Table of Content

23 March 2015, Volume 0 Issue 3

Previous Issue    Next Issue

An Improved Artificial Colony Algorithm for Real-time Path Planning of Mobile Robot

YIN Xia-hong, NI Jian-jun, WU Liu-ying

2015, 0(3):  1-4.  doi:10.3969/j.issn.1006-2475.2015.03.001

Asbtract ( 271 )  

References | Related Articles | Metrics

Path planning problem is one of the most important and challenging issues in mobile robot control field. In this paper, the grid method was used to model the robot working environment firstly. Then an improved artificial colony algorithm was put forward for robot path planning. In this algorithm, an adaptive search method was proposed in order to improve the convergence speed of the artificial colony algorithm and the elitist selection strategy was used to avoid the robot path planning falling into a local optimum. Experimental results show the feasibility and effectiveness of the proposed algorithm in path planning for mobile robots.


A Modified Particle Swarm Optimization Algorithm

XU Sheng-bing1, XIA Wen-jie1, FENG Ji-qiang2

2015, 0(3):  5-8,14.  doi:10.3969/j.issn.1006-2475.2015.03.002

Asbtract ( 208 )  

References | Related Articles | Metrics

In order to overcome the disadvantages of particle swarm optimization algorithm such as premature, bad searching capability, a Modified Particle Swarm Optimization (MPSO) algorithm was proposed. In this algorithm, when particles fall into the local extreme area, the particles, which gathered at the global optimal particle of the swarm, would have a mutation. Then test results of six complex benchmark functions and computer color matching model indicate that MPSO is superior to other two classic PSOs, which has high convergence precision, fast convergence speed and prevents particle premature effectively.


An Efficient Code Search Algorithm

LYU Fei

2015, 0(3):  9-14.  doi:10.3969/j.issn.1006-2475.2015.03.003

Asbtract ( 233 )  

References | Related Articles | Metrics

Over the years of software development, a vast amount of source code has been accumulated. At the same time, many code search tools have been developed to help programmers searching source code. As all these code search tools are not accurate enough, few developers use these tools to help them. In this paper, we propose an efficient code search algorithm, which can recognize the relation between APIs and queries, so it can improve the accuracy in code search. This paper also implements a C# code search tools based on the algorithm. At last, we did the experiment objectively and conducted a user study to evaluate the algorithm. The experiment results verify that the algorithm is efficient.


Support Vector Machine Based on Neighbor Edge Detection

WANG Xiu-hua, WU Li-fen

2015, 0(3):  15-19+25.  doi:10.3969/j.issn.1006-2475.2015.03.004

Asbtract ( 187 )  

References | Related Articles | Metrics

This paper presents a Support Vector Machine (SVM) method based on neighbor edge detection, called Support Vector Machine based on Neighbor Edge Detection (ED_SVM), in order to solve the problem that there is low training efficiency and it can not solve the large scale data mining problems of normal SVM， because it needs save, compute and solve the large kernel matrix. By dividing data and obtaining the mixed clusters, this method extracts the important samples near the approximate optimal hyperplane by introducing neighbor edge detection technology into the SVM training process, which have the most important support vector information. The new training samples set is constructed by these new important samples to keep the distribution feature of original support vectors and compress the size of training dataset. Then the normal SVM is trained on these new training samples and the good generalization performance can be obtained with high learning efficiency synchronously. The experiment results demonstrate that the proposed ED_SVM model can obtain the high learning efficiency and testing accuracy simultaneously.


Calculating Similarity of XML Documents by Weighted Pq-gram Algorithm

WANG Cheng-yong, DU Qing-wei, SUN Jing, SUN Zhen

2015, 0(3):  20-25.  doi:10.3969/j.issn.1006-2475.2015.03.005

Asbtract ( 192 )  

References | Related Articles | Metrics

Clustering for XML documents is an important method for efficiently managing XML documents, and calculating similarity of XML documents is the pivotal step. Pq-gram algorithm is an efficient method to solve the problem of calculating similarity of XML documents. However, it ignores that the nodes of XML documents are ordered. Based on the pq-gram algorithm, weighted pq-gram algorithm, in accordance with the structural characteristics of XML documents, sets weight for nodes, and sets weight for pq-grams based on the weight of nodes, then applies the weight to the method of calculating similarity of XML documents. Experimental results show that the weighted pq-gram algorithm describes the contribution of nodes better in the process of calculating similarity of XML documents, and improves the precision of calculating of XML documents.


Outliers Detection Algorithm Based on Density Division

WEI Long1, WANG Yong2

2015, 0(3):  26-32.  doi:10.3969/j.issn.1006-2475.2015.03.006

Asbtract ( 201 )  

References | Related Articles | Metrics

Most existing outliers detection algorithms need to input parameters manually, can’t detect the global and local outliers at the same time, and can’t deal with such problems as uneven density data effectively. This paper proposed an outliers detection algorithm DD-DBSCAN based on density division. The main innovation includes: 1) Define a new concept of Cluster Density according to the method of Minimum Spanning Tree, the entered data is divided into many clusters of different density. It can handle the data of uneven distribution density; 2) Adopting the idea of “divide and rule”, detect outliers from the division data respectively, make the algorithm be able to deal with the global and local outliers at the same time; 3) It can calculate the parameter value for each cluster automatically, makes the algorithm needs no longer human input parameters (Clustering Radius (Eps) Etc). Experiments on 2D simulated data sets and Iris real data sets, compared with DBSCAN algorithm, the results show that the proposed algorithm has higher precision and accuracy.

Parameters Optimization of Support Vector Machine Based on Improved Genetic Algorithm

WANG Qiong-yao1, HE You-quan2, PENG Xiao-ling1

2015, 0(3):  33-36.  doi:10.3969/j.issn.1006-2475.2015.03.007

Asbtract ( 185 )  

References | Related Articles | Metrics

 Over-study or under-study phenomenon sometimes happens, since nuclear parameters are chosen inappropriately in regression forecasting. The paper proposes a kind of support vector machine parameters optimization model based on improved genetic algorithm. By combining genetic algorithm with support vector machine algorithm, the model makes use of the principle of evolutionary of genetic algorithm to optimize penalty parameter, nuclear parameter and loss function at the same time, which are of great significance to support vector machine algorithm. Three sets of standard experiment data sets are selected as the test data set, and simulation test results are compared among the improved algorithm, genetic algorithm, particle swarm optimization algorithm and grid search algorithm. Experiment results show that the improved algorithm greatly improves the whole optimization ability of support vector machine algorithm.


Collaborative Filtering Recommendation Based on Optimization Euclidean Distance

CHEN Xiao-hui, GAO Yan

2015, 0(3):  37-40,47.  doi:10.3969/j.issn.1006-2475.2015.03.008

Asbtract ( 231 )  

References | Related Articles | Metrics

User evaluation data of items often are of the biodiversity and sparse characteristic in collaborative filtering recommendation system, the traditional similarity measurement algorithm cannot effectively find similar neighbors, this paper proposed a neighbor similarity computing algorithm based on optimized Euclidean distance. The algorithm introduced normalization and Jaccard similarity coefficient based on Euclidean distance calculation, and finally made the evaluation prediction and recommendation. The experiments result on typical dataset show that the algorithm can effectively improve the performance of collaborative filtering recommendation system.


Classifier Ensemble for Imbalanced Data Stream Classification Based on Accumulated Minorities

GUO Wen-feng1， WANG Yong2

2015, 0(3):  41-47.  doi:10.3969/j.issn.1006-2475.2015.03.009

Asbtract ( 203 )  

References | Related Articles | Metrics

 To solve the issue of over-fitting and not making full use of current data in existing methods of balancing imbalanced data stream, a method named EAMIDS for imbalanced data stream is proposed based on accumulated positive samples. In EAMIDS, positive samples from previous training chunks are accumulated to form the AP set which is used to balance the class distributions by making use of K nearest neighbors and Over-sampling technique. The ensemble classifier will be updated according to F-Measure when the number of the available base classifiers is greater than the fixed size of the ensemble classifier. Empirical study on both SEA dataset and SPH dataset shows that the proposed EAMIDS has substantial advantage over IDSL approach and SMOTE approach in prediction accuracy.


An Improved K-means Optimization Approach for Text Clustering

WANG Qiong

2015, 0(3):  48-51,56.  doi:10.3969/j.issn.1006-2475.2015.03.010

Asbtract ( 192 )  

References | Related Articles | Metrics

K-meansSC as an improved k-means optimization approach for text clustering is proposed. By means of processing of word segmentation, clustering document sets will be treated for extraction of main entry sets. Then the feature vectors of the document are respectively represented by Boolean function and TFIDF function, through the comparison of their respective strengths and weaknesses. Based on the entry set building support degrees matrix and confidence degrees matrix, similar degrees calculation formula can be defined, and under different clustering number conditions the formula and  other distance calculation formula of iteration number and error function of performance situation have been in detailed analysis. Experimental results shows that under certain conditions TFIDF function featuring document vector can effectively improve processing efficiency and clustering effectiveness.


Face Feature Extraction and Recognition Method for Multi-scale Local Structure-based Image Decomposition

FENG Xiang, YANG Jian, QIAN Jian-jun

2015, 0(3):  52-56.  doi:10.3969/j.issn.1006-2475.2015.03.011

Asbtract ( 186 )  

References | Related Articles | Metrics

In order to effectively extract the global and local features to improve the performance of face recognition, this paper presents a robust yet simple feature extraction method, called multi-scale image decomposition based on local structure. In the algorithm, the face image pyramid is first constructed through a multi-scale analysis. Then the local structural information by describing the relationship between the central macro-pixel and its neighbors for each level of the image pyramid is captured. In this way, one image is actually decomposed into a series of sub-images. Finally, all the structure images, after being down-sampled, are concatenated in one super-vector. Experimental results show that the proposed method is superior to some traditional methods such as Gabor, LBP and IDLS.


Binary Grid Domain Description Based on Morphology for One-class Classification

GAO Feng, QU Jian-ling, GUO Chao-ran, SUN Wen-zhu

2015, 0(3):  57-61.  doi:10.3969/j.issn.1006-2475.2015.03.012

Asbtract ( 117 )  

References | Related Articles | Metrics

 A one-class classifier of Binary Grid Domain Description (BGDD) based on morphology is proposed for solving unbalanced samples classification problems. In this method, the sample space is first divided into grids. Then, an approximate domain description can be obtained by putting samples into these grids. These grids are divided into object grids and background grids, the grid which contains at least one sample is defined as the object grid, while the grid without any sample is defined as the background grid. Next, morphological closing and opening operations are applied to object grids to obtain the domain description of the training samples. Experiments based on both artificial and real-world datasets were done and comparative experimental results were  present. Experimental results show that the BGDD classifier is an effective classification method for high classification accuracy and fast training speed.


Aircraft Attitude Estimation System Based on Improved Adaptive UKF Algorithm

ZHENG Jun-hui1， ZHANG Guo-ping2

2015, 0(3):  62-64,70.  doi: 10.3969/j.issn.1006-2475.2015.03.013

Asbtract ( 153 )  

References | Related Articles | Metrics

To address the aircraft attitude estimation problem, Unscented Kalman Filter (UKF) does not have the ability to overcome the disturbance. In order to obtain better estimation precision, an adaptive UKF is proposed. The algorithm can overcome the disturbance by introducing an adaptive factor to adjust the state gain matrix. The simulation results show the effectiveness of the proposed algorithm.


Multi-lane Mixed  Road Traffic Flow Based on Cellular Automata Model

QIU Fu-cheng1, LAN Shi-yong1,2, LI Yi1

2015, 0(3):  65-70.  doi:10.3969/j.issn.1006-2475.2015.03.014

Asbtract ( 170 )  

References | Related Articles | Metrics

The running of vehicle is influenced by the vehicles of the driveway and the front or the later vehicles of the adjacent lanes. In city traffic, the impact of mixing motor vehicles and non-motor vehicles on the running of vehicle is particularly complex. Through the research on the essential characteristics of mixed traffic flow, considering the “friction” and “blocking” interference of non-motor vehicle on the motor vehicles in lane changing rules and the acceleration and deceleration rules. The cellular automata (CA) model describing the mixed traffic is proposed, and numerical simulatings show the basic law of mixed traffic flow fundamental diagram when the non-motor vehicles and motor vehicles blended in different ratio.


PID Neural Network Optimizing Control Based on Particle Swarm Optimization in Paper Process

WU Xin-sheng

2015, 0(3):  71-74,79.  doi:10.3969/j.issn.1006-2475.2015.03.015

Asbtract ( 211 )  

References | Related Articles | Metrics

The optimal control of basis weight and moisture content in paper process with strong coupling, nonlinear and large time delay is difficult to achieve. To solve the problem, the optimal PID neural network controller by particle swarm optimization was adopted in the control system. Because the network structure was simple and a modified error back propagation algorithm with momentum factor was used, the learning speed was increased and the reaction time of the system became short. Particle swarm optimization was used to optimize the initial weights of PID neural network to avoid local optimization for obtaining better control accuracy. Simulation results show PID neural network optimizated by the network’s initial weights is of better adaptability, decoupling ability and robustness in the decoupling control of basis weight and moisture content. It is a new method for the control of basis weight and moisture content in paper process.


TENA Simulation Entities

SHI Jin-lin

2015, 0(3):  75-79.  doi: 10.3969/j.issn.1006-2475.2015.03.016

Asbtract ( 280 )  

References | Related Articles | Metrics

TENA is a test and training enabling architecture defined by the US Department of Defense project initiate FI2010, for facilitating the test and training range resource sector’s interoperability, reuse, and composability, and more efficiently using range resources, improving joint pilot training capability, reducing operating cost. Based on the analysis of the TENA meta-model and the object model, this paper studies the meaning of TENA simulation entities and composition, designs standard interface description for simulation entities, and simplifies the integration process of TENA object model that can quickly build simulation entities. The study can provide useful lessons for similar applications.


Design and Implementation of Mine Low-power Wireless Sensor Nodes

YANG Xue, TIAN Hong-xian

2015, 0(3):  80-83,107.  doi:10.3969/j.issn.1006-2475.2015.03.017

Asbtract ( 182 )  

References | Related Articles | Metrics

By applying MSP430 as a micro-controller unit, this paper focuses on the design of sensor nodes for a mine wireless roof separation monitoring system. The main work is to use a wireless wake mode (WOR) to periodically monitor the channel, collect the roof-shift data in real time, and send data to monitoring sub-station by RF communication module. Based on the analysis of low-power MSP430 principle and work mode, this article introduces the low power node design in detail from hardware design, work principle and communication mode, also its low power applications achievement. The wireless sensor nodes are of low power consumption and high reliability, so that maximize the use of limited energy, and have a high practical value in the harsh environment of mining.


Modern Uyghur Language Sentence Classification Tchnology

Azhar， Azragul， Yusup Abaydula

2015, 0(3):  84-87.  doi: 10.3969/j.issn.1006-2475.2015.03.018

Asbtract ( 167 )  

References | Related Articles | Metrics

The written text is usually composed of multiple sentences in natural language processing. And sentence boundary identification, sentence classification and analysis play an important role because the processing results have impact on the subsequent process, such as parsing and semantic analysis etc. This paper introduces an approach to classify the modern Uyghur sentences. Firstly， we introduce the construction rules of Uyghur sentence and the rules of automatic classification of simple and compound sentences. Secondly, we investigate the classification rules, principles, and algorithm for the Uyghur sentences. Then, the functional description, data description and system design and implementation processing description are given for the realized modern Uyghur sentence classification system. Finally, the experimental results are analyzed and the conclusions are represented.


Greenhouse Temperature and Humidity Controller Based on STC89C52

LONG Jian-ming, XIONG Gang, ZHANG Zheng-gang, HE Guo-rong, NIU Jia

2015, 0(3):  88-90,95.  doi: 10.3969/j.issn.1006-2475.2015.03.019

Asbtract ( 122 )  

References | Related Articles | Metrics

Against on low efficiency, high salary cost about greenhouse in some farmers and small agricultural enterprises, we designed a kind of controller which put single chip STC89C52 as core and combined temperature and humidity sensor, it had some functions such as real-time display, real-time control, out of limit alarming, parameter setting and so on. The test results show that the controller had simple operation, high stability, and better control performance. It can meet the requirement of temperature and humidity control and has better social and economic benefits.


Research and Application of Fast Polar Shape Descriptor

HU Yao-min

2015, 0(3):  91-95.  doi:10.3969/j.issn.1006-2475.2015.03.020

Asbtract ( 140 )  

References | Related Articles | Metrics

In real-time shape recognition based on video, the shape descriptor is desired to characterize the shape accurately, and can be extracted fast. To solve this problem, the fast polar shape descriptor (FPSD) is proposed for shape recognition. FPSD sampled one value at each angle frequency, and improved the sampling angle frequencies compared to Zernike Moments (ZM). Vehicle classification video was used for verifying experiments. When the FPSD is used as features for classification, experiments show that the performance of FPSD can be achieved even better than that of ZM, while the computational complexity is much lower than ZM.


Vehicle Gas Cylinder Supervision System Based on IOT

MA Yu-peng1,2, JIANG Tong-hai2

2015, 0(3):  96-100.  doi:10.3969/j.issn.1006-2475.2015.03.021

Asbtract ( 203 )  

References | Related Articles | Metrics

With the continuous growth of Xinjiang gas vehicles, vehicle cylinders leakage and the vehicle ignition accidents frequently occurred, causing casualties and bad social influence. To strengthen the gas cylinder safety management and implement dynamic monitoring become the problem that quality inspection departments have to face. In view of the traditional information methods can not effectively prevent the illegal modification and black cylinder, this paper proposed a vehicle gas cylinder supervision system based on Internet of things technology, relying on Tianshan Cloud computing platform, adopting J2EE multi-layer architecture and distributed deployment mode. The system supports a variety of communication protocols, adapts to the cable and wireless communication, serves in supervision institutions, filling mechanism, inspection mechanism, and modification mechanism. It realizes the effective supervision of real-time, dynamic, and the whole life cycle of the vehicle gas cylinder.


Building of Active Operational Case System Based on Case Analysis Pyramid Model

CAO Yi-feng, SHANG Hong-bin, CHEN Jie, BAO Yan-ping, SHEN Jing, LIU Xu, CHEN Xiao-wei

2015, 0(3):  101-107.  doi:10.3969/j.issn.1006-2475.2015.03.022

Asbtract ( 161 )  

References | Related Articles | Metrics

The traditional operational cases in the experience sharing and the disposal efficiency are relatively passive, in addition, the similar events in the operational field have a high probability of repetition and great differences in the characteristics of the disposal. For this situation, we present a case analysis pyramid model using library construction method, which is based on vectorization and hierarchical knowledge extraction, to complete the automatic changes from operating standards and events bases to case material library, then to case teaching library. Based on the pyramid model, through the introduction of improved predict analytical methods, case-based reasoning “5R” model and the “4S” case teaching model, this paper improved the related applications such as early warning and forecasting, decision supporting and case teaching, building active operational case system. Through the implementation of the whole system, practical results show that the proportion of similar events was declined year by year, event responsing time and handling time were shortened, and the quality of operational service was enhanced significantly.


Real-time Data Warehouse Accessing Technology Based on Dynamic Mirror Replication

MAO Ying-chi1,2, MIN Wei1, JIE Qing1, ZHU Li-li1

2015, 0(3):  108-112.  doi: 10.3969/j.issn.1006-2475.2015.03.023

Asbtract ( 148 )  

References | Related Articles | Metrics

Real-time data warehouse is one of important research fields in data management. Real-time data query and import can bring about the problem of query contention. Query contention will seriously affect the accuracy of query analysis. In this paper, combining an external dynamic storage area, a dynamic mirror replication technology was proposed to effectively solve the query contention problem; meanwhile, we improved the traditional ETL technology. Based on the TPC-H benchmark, the proposed dynamic mirror replication technology was evaluated. The experimental results demonstrate the query efficiency increased by 50%, and the query accuracy reached 93% on average.


Missing Value Estimation and Visualization of Meridian Datain Traditional Chinese Medicine

CHEN Jia-chang

2015, 0(3):  113-116.  doi:10.3969/j.issn.1006-2475.2015.03.024

Asbtract ( 189 )  

References | Related Articles | Metrics

With the development of Chinese culture, traditional Chinese medicine meridian is getting more and more attention and recognition. However, in the process of collecting of meridian data, it is inevitable that results data are missed which affects the further processing of the data. Also, because the meridian data is of high dimensions, long timing and so on, traditional data mining methods are difficult to deal with it. This article proposes a missing meridian data visualization analysis framework. The experimental results show that the visual analysis framework effectively compensates for the loss of missing data, and through visual display, it can dig out some hidden rules in the meridian data.


Authentication Mechanism in Mobile Routing System

ZHAO Jiang-yun, DONG Ping, GAO De-yun

2015, 0(3):  117-121,126.  doi:10.3969/j.issn.1006-2475.2015.03.025

Asbtract ( 163 )  

References | Related Articles | Metrics

The mobile routing system based on tunnel technology provides a new solution to solve the communication problem between train and ground. In order to protect the system security, making the user’s online behavior controllable and manageable, this paper proposes the needs of authentication mechanism. By analyzing the principle of the routing system, software and hardware conditions, this paper gave an efficient and easy implementation to authentication mechanism, and realized it with Netfilter/ipatbles, PHP, MySQL, and other tools under Linux. At last, a topology had been built and the authentication mechanism was tested, clearly showing that the mechanism could work well and satisfy the design needs.


Design of Low Power Node on Wireless Sensor Network Based on Contiki OS

LIU Jia-yu, GAO De-yun

2015, 0(3):  122-126.  doi:10.3969/j.issn.1006-2475.2015.03.026

Asbtract ( 219 )  

References | Related Articles | Metrics

 Power efficiency should be the primary concern when designing wireless sensor network because nodes are energy constrained. Through low power hardware selection, a WSN node was proposed based on ultra-low power MCU MSP430F5438A and IEEE 802.15.4 transceiver AT86RF231. The use of Contiki OS eliminated foreground/background software design pattern or finite state machine, and developers could use modularized and thread-like code style in software designing. Test platform was built to perform multi-point information gathering experiment and static power measurement. These tests show practical use and low-power feature of the node.
